As large language models are increasingly deployed as tool-augmented legal agents, they introduce agentic hallucinations where tool-call and reasoning errors cascade into fabricated holdings and misci...
arXiv:2606. 18021v1 Announce Type: new Abstract: AI systems deployed in legal workflows hallucinate at rates that aggregate metrics report at ~52%, but this average conceals where errors concentrate and in which direction they run, leaving compliance officers without an actionable signal for trustworthy deployment.
By Lalit Yadav, Akshaj Gurugubelli
arXiv:2601.22984v3 Announce Type: replace
Abstract: Diagnosing failure patterns in Deep Research Agents (DRAs) remains a critical challenge. Existing benchmarks predominantly rely on end-to-end evalu...
By Yuhao Zhan, Tianyu Fan, Linxuan Huang, Zirui Guo, Chao Huang
The paper argues that hallucinations by legal language models should be judged as failures of legal warrant rather than mere factual or citation errors. It defines claim-authority warrant as a context-sensitive relationship between a legal claim and applicable, current authority, and proposes that evaluating warrant can uncover failures missed by traditional accuracy or citation metrics. The authors outline a pilot study, benchmark specifications, and a research agenda to assess whether legal AI systems’ claims are properly licensed by law.
By Maksym Taranukhin, Vered Shwartz
arXiv:2608. 14210v1 Announce Type: cross Abstract: Hallucination is a major challenge for retrieval-augmented generation (RAG) systems in the legal domain, where ungrounded answers can lead to serious consequences.
By Souvick Das, Sallam Abualhaija, Domenico Bianculli
Legal Research Bench (LRB) is a new benchmark comprising 413 open-ended U.S. legal research questions, each paired with a gold answer, supporting authorities, and a binary grading rubric. The study evaluates thirteen advanced language‑model agents using web search, case‑law search, page parsing, and retrieval tools, scoring responses only when all required criteria are met and cited authorities verify. Results show that even the best model, Claude Opus 4.8, achieves full correctness on only 42.9% of questions, with performance varying by legal area and task complexity, and no clear link between more tool calls or inference cost and higher accuracy.
By Katrina Drozdov, Oliver Chen, Langston Nashold, Rayan Krishnan